發(fā)布時間：2018-01-19 23:59

  本文關鍵詞： 天然氣管網 連通可靠性 全生命周期成本 粒子群優(yōu)化算法　出處：《西南石油大學》2017年碩士論文　論文類型：學位論文

【摘要】：天然氣作為高效、經濟的清潔能源,隨著我國國民經濟的飛速發(fā)展,對其需求也日益增加。天然氣管網作為輸送能源的重要樞紐,已經成為國家能源供應的不可或缺的一環(huán)。對于大型天然氣管網來說,從全局、全生命、全系統(tǒng)的高度對天然氣管網進行規(guī)劃管理,是發(fā)展的必然趨勢。因此,本文結合可靠性理論,對天然氣管網連通可靠性、天然氣管網全生命周期費用優(yōu)化進行了研究。(1)搭建管道系統(tǒng)全生命周期成本分析的框架體系,為管道系統(tǒng)全生命周期成本費用優(yōu)化提供可以參考的依據。將全生命周期成本分析分為4個過程:確定范圍、數據采集和費用結構分解、分析與建模以及報告和決策。上述4個過程在實現全生命周期成本費用最小的目標之前,可以不斷重復,直到達到目標要求。(2)天然氣管網目標可靠度全生命周期優(yōu)化決策可以分為兩個階段:管道系統(tǒng)目標設計可靠度全生命周期優(yōu)化決策階段和天然氣管網目標運行可靠度全生命周期優(yōu)化決策階段,選取管道結構和站場作為兩個階段的優(yōu)化對象。針對管道系統(tǒng)全生命周期優(yōu)化決策,提出了以成本造價函數、維護維修費用函數、失效損失期望函數三者之和為目標函數的管道系統(tǒng)全生命周期成本費用函數,并分別給出了成本造價函數、維護維修費用函數、失效損失期望函數三者與設計可靠度之間的關系。根據工程實際情況,以可靠度約束和投資約束作為約束條件,建立了兩種管道系統(tǒng)全生命周期成本優(yōu)化決策模型。(3)依據圖論,將天然氣管網經過簡化后轉化為抽象的圖一拓撲結構,并給出單氣源和多氣源天然氣管網系統(tǒng)的連通可靠性指標及其邏輯表達式。通過將最小路集轉化為矩陣,進行向量與矩陣的特殊運算,連通可靠度的邏輯表達式轉化為解析表達式,得以計算連通可靠度。(4)針對天然氣管網,提出了以維護維修費用函數、失效損失期望函數二者之和為目標函數的天然氣管網全生命周期成本費用函數,并分別給出了維護維修費用函數、失效損失期望函數二者與運行可靠度之間的關系,以氣源和門站的連通可靠度作為約束條件,構建了天然氣管網運行可靠度優(yōu)化決策模型。(5)對于復雜的連續(xù)函數,采用粒子群算法對其進行優(yōu)化。經過計算發(fā)現,慣性權重0.4≤ω≤0.9、加速常數1.5≤c1≤2.0以及1.5≤c2≤2.0時,粒子群算法的精度能夠滿足要求。最后,為進一步提高算法的精度,在粒子群算法中引入收縮因子,并根據迭代步數對慣性權重、加速常數動態(tài)取值。
[Abstract]:Natural gas as an efficient and economic clean energy, with the rapid development of the national economy in China, the demand for natural gas pipe network as an important hub for energy transmission is also increasing day by day. Has become an indispensable part of national energy supply. For large-scale natural gas pipeline network, from the overall situation, the whole life, the whole system height of the natural gas pipeline network planning and management, is the inevitable trend of development. Based on the reliability theory, this paper studies the reliability of natural gas pipeline network and the optimization of the whole life cycle cost of natural gas pipeline network, and establishes the framework system of the whole life cycle cost analysis of pipeline system. The whole life cycle cost analysis is divided into four processes: determination range, data acquisition and cost structure decomposition. Analysis and modeling, as well as reporting and decision-making. These four processes can be repeated until the goal of minimizing the cost of the entire life cycle is achieved. Until the target is met. The whole life cycle optimization decision of target reliability of natural gas network can be divided into two stages:. The whole life cycle optimization stage of pipeline system reliability and the whole life cycle optimization stage of natural gas pipeline network target operation reliability. The pipeline structure and the station yard are selected as the optimization objects in two stages. Aiming at the whole life cycle optimization decision of the pipeline system, the cost cost function and the maintenance cost function are put forward. The life cycle cost function of pipeline system with the sum of the expected function of failure loss is the objective function, and the cost cost function and maintenance cost function are given respectively. The relationship between the expected function of failure loss and the design reliability. According to the actual situation of the project, the constraint of reliability and investment is taken as the constraint condition. Based on graph theory, two kinds of pipeline systems are transformed into abstract graph-topology. The connected reliability index and its logical expression of single gas source and multi gas source natural gas pipeline network system are given. By transforming the minimum path set into matrix, the special operation of vector and matrix is carried out. The logical expression of the connected reliability is transformed into an analytical expression, and the connected reliability can be calculated. 4) in view of the natural gas network, the maintenance cost function is put forward. The whole life cycle cost function of natural gas pipeline network with the sum of the expected failure loss function and the target function is given, and the maintenance cost function is given respectively. The relationship between the expected function of failure loss and the operational reliability is restricted by the connectivity reliability of the gas source and gate station. The optimization decision model of running reliability of natural gas network is constructed. (5) for the complex continuous function, particle swarm optimization algorithm is used to optimize it. It is found that the inertia weight is 0.4 鈮，

本文編號：1446024